Evidence Brief: Effects of Small Hospital Closure on ... · 2003;9(2):59-71. 5. Bazzoli GJ, Lee W,...

Department of Veterans AffairsHealth Services Research & Development Service Evidence-based Synthesis Program

Evidence Brief: Effects of Small Hospital Closure on Patient Health OutcomesSUPPLEMENTAL MATERIALS

June 2013

Prepared for:Department of Veterans AffairsVeterans Health AdministrationQuality Enhancement Research InitiativeHealth Services Research & Development ServiceWashington, DC 20420

Prepared by:Evidence-based Synthesis Program (ESP) Coordinating CenterPortland VA Medical CenterPortland, ORMark Helfand, MD, MPH, MS, Director

Investigators:Principal Investigator:

Susan Carson, MPH

Contributing Investigators: Kim Peterson, MSLinda Humphrey, MD, MPHMark Helfand, MD, MPH, MS

http://www.hsrd.research.va.gov/

http://www.hsrd.research.va.gov/publications/esp/

http://www.queri.research.va.gov/

Effects of Small Hospital Closure on PatientHealth Outcomes: Supplemental Materials Evidence-based Synthesis Program

Page i9CONTENTS 34

TABLE OF CONTENTS

SEARCH STRATEGIES .................................................................................................................................. 1

LIST OF EXCLUDED STUDIES ................................................................................................................. 2

EVIDENCE TABLESData Abstraction of Included Studies ............................................................................................................ 6Quality Assessment of Included Studies ..................................................................................................... 10


Page 19CONTENTS 34

SEARCH STRATEGIESPubMed Search (March 6th 2013)

Search Query#28 Search (#27) AND #25#27 Search (closing[Title]) OR closure[Title]#25 Search (#24) AND #14#24 Search (#22) OR #23#23 Search (hospital[Title]) OR hospitals[Title]#22 Search (((“Hospital Bed Capacity, under 100”[Mesh]) OR ( “Hospitals, Low-

Volume”[Mesh] OR “Hospitals, Satellite”[Mesh] OR “Hospitals, Veterans”[Mesh] )) OR “Hospitals, Public”[Mesh]) OR “Hospitals, Rural”[Mesh]

#14 Search (#13) OR #12#13 Search downsizing[Title]#12 Search (“Health Facility Closure”[Mesh]) OR “Health Services Accessibility”[Mesh]

Ovid MEDLINE and OLDMEDLINE (1946 to March Week 4 2013)Search Query#1 *Health Facility Closure/#2 *Health Services Accessibility/#3 1 and 2#4 Limit 3 to English

PubMed related records searches on the following articles (March 6th 2013):

1. Bindman AB, Keane D, Lurie N. A public hospital closes. Impact on patients’ access to care and health status. JAMA. Dec 12 1990;264(22):2899-2904.

2. Buchmueller TC, Jacobson M, Wold C. How far to the hospital? The effect of hospital closures on access to care. J Health Econ. 2006;25(4):740-761.

Forward citation searching of the following articles via Google Scholar and Scopus (March 7, 2013)

1. Bindman AB, Keane D, Lurie N. A public hospital closes. Impact on patients’ access to care and health status. JAMA. Dec 12 1990;264(22):2899-2904.

2. Buchmueller TC, Jacobson M, Wold C. How far to the hospital? The effect of hospital closures on access to care. J Health Econ. 2006;25(4):740-761.

3. Hemmelgarn BR, Ghali WA, Quan H. A case study of hospital closure and centralization of coronary revascularization procedures. CMAJ. 2001;164(10):1431-1435.

4. Liu L, Hader J, Brossart B, White R, Lewis S. Impact of rural hospital closures in Saskatchewan, Canada. Social Science and Medicine. 2001;52(12):1793-1804.

http://www.ncbi.nlm.nih.gov/pubmed/advanced










Page 29CONTENTS 34

LIST OF EXCLUDED STUDIES (No original data or no health outcomes reported)1. AIexander JA, Succi MJ. State legislation and policy affecting rural hospital conversion

and closure. J Rural Health. 1996;12(5):410-422.

2. Aldrich RR. The merger and closing of an urban hospital. Case Stud Health Adm. 1978;1:269-274.

3. BarnettJR.Rationalisinghospitalservices:reflectionsonhospitalrestructuringanditsimpacts in New Zealand. N Z Geog. 2000;56(1):5-21.

4. Barnett R, Barnett P. “If you want to sit on your butts you’ll get nothing!” Community activism in response to threats of rural hospital closure in southern New Zealand. Health Place. 2003;9(2):59-71.

5. Bazzoli GJ, Lee W, Hsieh H-M, Mobley LR. The effects of safety net hospital closures and conversions on patient travel distance to hospital services. Health Serv Res. 2012;47(1 PART 1):129-150.

6. Berrey PN. Increase in acute admissions and deaths after closing a geriatric day hospital. Br Med J (Clin Res Ed). Jan 18 1986;292(6514):176-178.

7. Bolton P, Thompson L. The reasons for, and lessons learned from, the closure of the Canterbury GP After-Hours Service. Aust Health Rev. 2001;24(3):66.

8. Bovbjerg RR, Marsteller JA, Ullman FC. Health Care for the Poor and Uninsured after a Public Hospital’s Closure or Conversion. 2000.

9. Brownell MD, Shapiro E, Roos NP. Re “A critique of an evaluation of the impact of hospital bed closures in Winnipeg, Canada: lessons to be learned from evaluation research methods” by Evelyn Vingilis and Jacquelyn Burkell. J Public Health Policy. 1997;18(4):469-471; author reply 472-464.

10. Burda D. Hospital closures aren’t closing off access to care. Mod Healthc. 1992;22(27):18-20.

11. Burkey ML, Bhadury J, Eiselt HA. A location-based comparison of health care services infourU.S.stateswithefficiencyandequity.Socio-Economic Planning Sciences. 2012;46(2):157-163.

12. CappsC,DranoveD,LindroothRC.Hospitalclosureandeconomicefficiency.J Health Econ. 2010;29(1):87-109.

13. Caro FG, Glickman LL, Ingegneri D, Porell F, Stern AL, Verma K. The impact of the closing of three Massachusetts public chronic disease hospitals: a multidimensional perspective. J Community Health. Jun 1997;22(3):155-174.

14. Chatwin AL, Miller M, Asensio J, Kerstein MD. Cause of temporary closure of an inner-city trauma center. Am Surg. 1995;61(12):1102-1104.

15. Drain M, Godkin L, Valentine S. Examining closure rates of rural hospitals: an assessment of a strategic taxonomy. Health Care Manage Rev. 2001;26(4):27-51.


Page 39CONTENTS 34

16. Fleming ST, Williamson Jr HA, Hicks LL, Rife I. Rural hospital closures and access to services. Hosp Health Serv Adm. 1995;40(2):247.

17. Goddard K, Burns T, Catty J. The impact of day hospital closure on social networks, clinical status, and service use: a naturalistic experiment. Community Ment Health J. Jun 2004;40(3):223-234.

18. Greene J. Study says closure decreased access. Mod Healthc. 1990;20(50):2.

19. HammondL.TheclosureofMelbourne’sFairfieldHospital.Aust N Z J Public Health. Jun 1996;20(3):230-231.

20. Harmata R, Bogue RJ. Conditions affecting rural hospital specialization, conversion, and closure: a case-based analysis of threat and change. J Rural Health. 1997;13(2):152-163.

21. Hart LG, Pirani MJ, Rosenblatt RA. Causes and consequences of rural small hospital closures from the perspectives of mayors. J Rural Health. 1991;7(3):222–245.

22. Holmes GM, Slifkin RT, Randolph RK, Poley S. The effect of rural hospital closures on community economic health. Health Serv Res. 2006;41(2):467-485.

23. Hsia RY-J, Shen Y-C. Rising closures of hospital trauma centers disproportionately burden vulnerable populations. Health Aff (Millwood). 2011;30(10):1912-1920.

24. James AM. Closing rural hospitals in Saskatchewan: On the road to wellness? Soc Sci Med. 1999;49(8):1021-1034.

25. JervisKJ,GoldbergGM,CuttingAC.Inner-cityhospitalclosures:financialdecisionorimpediment to access? J Health Care Finance. 2012;38(3):22-39.

26. Jervis KJ, Goldberg GM, Cutting AC. Rights, needs, and equality of opportunity for health care:Afinancialanalysisofmorality.Accounting and the Public Interest. 2012;12(1):62-86.

27. Klepser DG, Xu L, Ullrich F, Mueller KJ. Trends in community pharmacy counts and closures before and after the implementation of Medicare Part D. J Rural Health. 2011;27(2):168-175.

28. Lorch SA, Srinivas SK, Ahlberg C, Small DS. The impact of obstetric unit closures on maternal and infant pregnancy outcomes. Health Serv Res. 2012.

29. McCarthy JF, Blow FC, Valenstein M, et al. Veterans Affairs health system and mental health treatment retention among patients with serious mental illness: evaluating accessibility and availability barriers. Health Serv Res. 2007;42(3 I):1042-1060.

30. McKay NL, Coventry JA. Rural hospital closures. Determinants of conversion to an alternative health care facility. Med Care. 1993;31(2):130.

31. McKay NL, Coventry JA. Access implications of rural hospital closures and conversions. Hospital and Health Services Administration. 1995;40(2):227-246.

32. McKee M. What are the lessons learnt by countries that have had dramatic reductions of their hospital bed capacity http://www.isophp.ru/index.php?razdID=23&articleID=40&l=en. Health evidence network, WHO Regional Office for Europe. 2003.


Page 49CONTENTS 34

33. McKee M. Reducing hospital beds: what are the lessons to be learned? WHO Regional OfficeforEurope.2004.

34. McLafferty S. Neighborhood characteristics and hospital closures: a comparison of the public, private and voluntary hospital systems. Soc Sci Med. 1982;16(19):1667-1674.

35. McLafferty S. Predicting the effect of hospital closure on hospital utilization patterns. Soc Sci Med. 1988;27(3):255-262.

36. McNamara PE. Welfare effects of rural hospitial closures: a nested logit analysis of the demand for rural hospital services. Am J Agr Econ. 1999;81(3):686-691.

37. Mundy CM, Gilcreast DM. Organizational downsizing and closure in two U.S. Army hospitals. Mil Med. 1994;159(3):224.

38. Muus KJ, Ludtke RL, Gibbens B. Community perceptions of rural hospital closure. J Community Health. 1995;20(1):65–73.

39. Netten A, Darton R, Williams J. Nursing home closures: effects on capacity and reasons for closure. Age and ageing. 2003;32(3):332-337.

40. Petrucka PM, Wagner PS. Community perception of rural hospital conversion/closure: re-conceptualising as a critical incident. Aust J Rural Health. 2003;11(5):249-253.

41. Pugno PA, Kortum J, Beehler R, Cox F, Albright H, Anderson PA. The closing of Shasta General Hospital. JAMA: The Journal of the American Medical Association. 1991;265(14):1827–1827.

42. Reif SS, DesHarnais S, Bernard S. Community perceptions of the effects of rural hospital closure on access to care. J Rural Health. 1999;15(2):202-209.

43. Rohrer JE. Closing rural hospitals: reducing “institutional bias” or denial of access? J Public Health Policy. 1989;10(3):353-358.

44. Romero D, Kwan A, Nestler S, Cohen N. Impact of the closure of a large urban medical center: a quantitative assessment (part II). J Community Health. Oct 2012;37(5):995-1005.

45. Romero D, Kwan A, Swearingen J, Nestler S, Cohen N. Impact of the closure of a large urban medical center: a qualitative assessment (Part I). J Community Health. 2012;37(5):982-994.

46. RossbarnettJ.Rationalisinghospitalservices:reflectionsonhospitalrestructuringanditsimpacts in New Zealand. N Z Geog. 2008;56(1):5–21.

47. Sager A. The Proposed Closing of Rancho Los Amigos National Rehabilitation Center Endangers the Health of Disabled Medi-Cal Patients Who Reside in Los Angeles County. 2003.

48. Sager A, Socolar D. Closing Hospitals in New York State Won’t Save Money But Will Harm Access to Health Care. 2006.

49. Samuels S, Cunningham JP, Choi C. The impact of hospital closures on travel time to hospitals. Inquiry. 1991:194-199.


Page 59CONTENTS 34

50. Taylor J, Zweig S, Williamson H, Lawhorne L, Wright H. Loss of a rural hospital obstetric unit: a case study. J Rural Health. Oct 1989;5(4):343-352.

51. Walker KO, Clarke R, Ryan G, Brown AF. Effect of closure of a local safety-net hospital on primary care physicians’ perceptions of their role in patient care. Ann Fam Med. 2011;9(6):496-503.

52. Walker KO, Leng M, Liang L-J, et al. Increased patient delays in care after the closure of Martin Luther King hospital: implications for monitoring health system changes. Ethn Dis. 2011;21(3):356-360.

53. Wu VY. The price effect of hospital closures. Inquiry. 2008;45(3):280-292.

Effects of Small Hospital Closure on Patient Health Outcomes: Supplemental Materials Evidence-based Synthesis Program

69CONTENTS 34

EVIDENCE TABLES

Data Abstraction of Included StudiesAuthor Study design Closed hospital Comparator hospital Patient Confounders controlled Year N characteristics characteristics Time period characteristics Health outcomes Data sources for in analysis ResultsBindman Pre-Post Shasta County San Luis Obispo County Closed 1987 Mean age=49 years Self-reported health Patient surveys at Age, sex, race, work status, Mortality: Shasta=4% vs. SLO=3%, 1990 N=219 patients from General Hospital (SLO) 1987-1988 38% male status via the Medical baseline (just after insurance status, number P>0.10

closed hospital; 195 73 beds Public hospital in a 88% White Outcomes Study Short hospital closed) and of chronic conditions, and MOS-SF: Shasta patients had worse from comparator Acute care semirural county in 32% Medicaid Form (MOS-SF): pain, 1 year later. baseline response. Health Perception (P<0.05), Social hospital Public county hospital central California with Mean # chronic mental health, physical Function (P<0.05), Role Function

Semirural area of comparable urban and illnesses: 3.5 function, health (P<0.05), and Pain (P<0.01). But, Northern California rural distribution, total perception, social no significant difference in Physical

population, physician function, role function. Function or Mental Health.to population ratio, Mortality information ethnic mix, and median from family members household income. and Vital Statistics

Registry.

Brownell Pre-Post Downsizing number NA 1989-1995 Mortality: Patients Mortality 30, 60, and Population-based Age and sex Mortality rates within 30, 60 and 90 1999 N NR of acute beds in hospitalized for acute 90 days from hospital approach used days post-discharge: No statistically

Winnipeg hospitals MI, hip fracture, or discharge significant increases Total acute beds at cancer surgery For the subgroup of elderly patients, start of fiscal year 30-day readmission no increases across the years in (#/% decrease at year- Readmissions: Patients mortality. end): treated for 12 medical, 1991=3042 (N/A) surgical, and obstetric 30-Day Readmission Rates: Only 1992=3013 (29/1.0%) categories significant increase is for cesarean 1993=2707 section in 1995. (306/10.2%) 1994=2498 (209/7.7%) 1995=2460 (38/1.5%) 1996=2380 (80/3.3%)

Buchmueller Ecological/Cross Hospitals in the Los Effect of hospital closures 1997-2003 Mean age=44 years Health status, mortality, LA County Health For health outcomes (from Closures may induce some survival 2006 Sectional Angeles region through their effect on 56% Male mortality from heart Surveys for health surveys): income, health benefit:

N=23503 distance from nearest 56% White attacks and unintentional status and zip code insurance coverage, health % change in deaths due to a mile hospital to outcomes. Household income injuries sustained at (geographic area); status, neighborhood increase in distance to hospital Compared changes in >75,000 US$: 24% home. death reports from characteristics such as related to closures (per zip code year/health among individuals 61% Private Insured California DHS number of community health zip code specific time trends): in areas where hospitals Diabetes: 6% clinics in a zip code and city- AMI: -11.0%/-22.7% closed to changes Heart Disease: 6% level unemployment rates. Unintentional injuries, home: among otherwise similar Mean BMI=24.1 For deaths: controls for total -52.7%/-60.8% individuals in areas Mean self-assessed deaths, deaths by homicide Chronic heart disease: -2.37%/ where the availability of health (1=excellent, and age distribution of -8.13% hospital services remained 5=poor): 2.34 deaths, number of health All cancer: 6.28/9.09constant. clinics in a zip code year.


79CONTENTS 34

Author Year

Study design N

Closed hospital characteristics

Comparator hospital characteristics Time period

Patient characteristics Health outcomes Data sources

Confounders controlled for in analysis Results

Curtis2005

Time series?Sample sizes for within vs. outside St. John’s (1995-96/1998-99/2000-01):Acute myocardial infarction (AMI): 202/284 vs. 271/274/280Community acquired pneumonia(CAP): 226/336/264 vs. 108/122/110Cerebrovascular accident (CVA): 241/274/175 vs. 85/109/116

Implementation of program-based management, closed hospitals and aggregated services both within and outside of the largest health board of Newfoundland and Labrador (St. John’s region)

NA Regionaliza-tion occurred between 1995 and 1997. Evaluated outcomes at the beginning (1995/96), during (1998/99) and shortly after (2000/01) restructuring.

AMI Mean age 68 years60% male32% diabetes16% CHF

CAPMean age 68 years56% male21% diabetes71% Pneumonia Severity Index Class III-V

CVAMean age 71 years56% male

In-hospital mortality Data from medical records abstracted by trained research nurses

None In-hospital mortality rate for hospitals inside vs. outside of St. John’s Region (1995-96/1998-99/2000-01): AMI: 14%/12%/NR, P=NS vs. 10%/15%/9%, P=NSCAP: 9%/11%/10%, P=NS vs. 12%/6%/9%, P=NSCVA: 22%/20%/20%, P=NS vs. 25%/22%/18%, P=NS

Hemmelgarn 2001

Pre-Post Before closure: 1053 CABG, 2340 PTCA patients After closure: 1529 CABG, 3099 PTCA patients

Large city hospital in Calgary

After closure, coronary revascularization procedures were amalgamated from 2 facilities into a single facility, including the addition of 2 new coronary catheterization labs and an increased number of postoperative ward beds and monitored step-down beds at the single facility.

July 1994-March 1998 (21 months preceding and 24 months following the March 1996 closure of the hospital)

% patients undergoing CABG or PTCA: Age >65 years: 59%/43% Male: 78%/74% Unstable angina: 50%/36% Recent MI: 24%/40% Chronic pulmonary disease: 16%/7% Diabetes: 16%/14% Congestive heart failure: 22%/8%

For CABG and PTCA, number of discharges per month, burden of comorbidity, length of hospital stay and in-hospital mortality

Calgary Regional Health Authority administrative hospital discharge data

Age, sex, urgency of admission, prior procedures, and comorbidities

Adjusted mean LOS CABG patients: Before closure: 16.1 days After closure: 14.9 days P=0.004 PTCA patients Before closure: 5.1 days After closure: 4.1 days P<0.001 Adjusted in-hospital rates of death CABG patients: Before closure: 4.6% After closure: 4.3% P=0.67 PTCA patients Before closure: 1.8% After closure: 1.9% P=0.83

Hsia2012

Cohort N=785,385 (of which 67,577 experienced an increase in distance to their nearest ED)

Nonfederal hospitals in California (emergency departments only).All admissions for acute MI, stroke, sepsis, and asthma or COPD

Calculated distance from each patient’s home to the nearest ED. Compared patients with an increase in distance to the nearest ED during the time period to those without

1999-2009 % patients in age category (years): 18-44=7%, 45-64=26%, 65-74=21%, 75-84= 28%, >85=18% 46% male 66% Non-Hispanic White15% Hispanic65% Medicare59% Hypertension24% Diabetes

Inpatient mortality Patient-level data from the California Office of Statewide Health and Planning Development Patient Discharge Data.Use data from state to document which hospitals had ED closures.Data on hospital-level characteristics from hospital financial and use reports.

Age, race or ethnicity, sex, and insurance status, comorbidity index to adjust for baseline mortality risk, case mix of each hospital to control for hospital-level differences in patient acuity.

Inhospital mortality for all time-sensitive conditions of patients experiencing stepwise increasing distances to their nearest ED compared with those having no change or a decrease, adjusted OR (95% CI):Increase of <2 miles: 1.04 (0.99 to 1.09) Increase of 2-5 miles: 1.03 (0.91 to 1.16)Increase of >5 miles: 1.09 (0.95 to 1.26)

For cause-specific inhospital mortality, only significant increase was for stroke for patients who experienced an increase of >5 miles: 1.22 (1.02 to 1.47)

Only 8.6% of patients experienced an increase in distance, and among them the median distance was 0.8 miles (range 0.1 to 33.4 miles).


89CONTENTS 34

Author Year

Study design N





Liu 2001

Pre-Post N=47 communities in the closure group with a total population of over 56,000 per year

Saskatchewan communities in which hospitals funded for <8 beds were closed in 1993. Unit of analysis was the communities, not hospital.

Compared closure communities to: 1) rural communities that never had a hospital, 2) rural communities that still have a hospital, and 3) the rest of Saskatchewan

1990-1996 NR Mortality: overall, premature, and cause-specific (acute MI, motor vehicle injury, and stroke) Hospital use (number of people hospitalized, not number of discharges) Self-reported health status

Saskatchewan Ministry of Health administrative data. Patient surveys and interviews for health status.

Age and sex All-cause mortality per 100,000 people, 1990-92/1993-96 (change): Communities affected by 1993 hospital closures: 803.1/754.4 (-48.7) Communities still with small hospitals: 853.3/833.1 (-20.1) Communities that never had a hospital: 684.9/651.1 (-33.8) Rest of Saskatchewan: 788.3/774.8 (-13.5)

No increases in cause-specific mortality.

Closures did not adversely affect health status of residents. Mortality rates declined throughout the province, with largest decreases in the closure communities. Residents self-reported that closures did not adversely affect their health.

Rosenbach 1995

Time series11 hospitals (3 publicly owned, 4 private non-profits, 4 for-profit)

Rural hospitals in 6 US states 6 had fewer than 50 beds, only 1 had more than 100 beds.Medicare-certified, short-term, general acute-care facilities which closed in either 1986 or 1987. Closure defined as the cessation of inpatient care. Hospitals that converted from acute care to non acute-care status were included, as were mergers that resulted in a conversion to a non-acute-care facility.

Compared outcome measures in the year before closure, the year of closure, and the 2 years following closure. Comparators were Medicare beneficiaries in areas that had no hospital closures, openings, mergers, or conversions during the study period; and those in areas in which there were no hospitals during the entire study period.

1985-1989 NR Mortality Age and sex Number of deaths per 1,000 Medicare beneficiaries for one year before closure/year of closure/1-year post-closure/2-years post-closure/percent change:Closure: 36.4/38.4/37.6/38.1/ 4.7%No closure: 36.5/36.2/35.9/24.9/ -4.4%No hospital: 33.7/36.7/35.3/34.7/3.0%None of the 4-year trends was statistically significant. Profile analysis shows no significant effects from the closure.


99CONTENTS 34

Author Year

Study design N





Shen2012

Ecological/Cross Sectional N=1.49 million patient-year observations

ED’s in the U.S. Compared effect of ED closures through their effect on driving time to closest ED between 2 groups: 1) people living in zip codes with no increase in driving time and 2) people in zip codes with <10, 10 to 30, or >30 minute increases in driving time.

1995-2005 Acute myocardial infarction: 51% female Mean age 78.54 years 89% White Peripheral vascular disease: 8% Chronic pulmonary disease: 22% Dementia: 4% Chronic renal failure: 2% Diabetes: 27% 76% urban location (range, 17% to 75%) No. of hospitals within 10-mile radius: 2.70 (range, 1.03 to 4.22) Government hospitals: 12%

30-day to 1-year mortality rates

ED availability from American Hospital Association annual surveys Patients’ AMI diagnoses, mailing ZIP codes and outcomes from Medicare claims

Subgroup analyses based on differences in baseline access to hospitals

Change in mortality rate at 7 days, 30 days, 180 days, and 1 year by increased drive time for whole sample: <10 min: -0.0002, 0.0029, 0.0046, 0.0061 (P<0.10), 0.0037 10-30 min: -0.0063, -0.0098, -0.0061, -0.0026, -0.0072 >30 min: 0.0172, 0.0123, 0.0258, 0.0449 (P<0.10), 0.0565 (P<0.05) Closest hospital has catheterization lab: -0.0046 (P<0.05), -0.0047 (P<0.05), -0.0034, -0.0026, -0.0036 Subgroup analyses: Patients with ≤2 hospitals within 10-mile radius: Significant increases in mortality across all time points when driving increased by <10 minutes, but not for other groups with higher increases in driving time. But, increases in mortality are transitory, peaking at the time period of 1 year before to 1 year after change and then decreasing beyond initial 3-year window.


109CONTENTS 34

Quality Assessment of Included StudiesDid the article

Did the study maintain Did the study attempt to Were the groups use accurate Were outcome comparable Was there enroll all (or a random comparable at methods for assessors and/ groups (report acceptable sample of) patients baseline on key ascertaining or data analysts attrition, differential loss to Were outcomes pre-meeting inclusion criteria, prognostic factors exposures blinded to the contamination, Did the study perform follow-up and or specified and defined, and

Author or a random sample (e.g., by restriction and potential exposure being adherence, and appropriate statistical analyses overall high loss to ascertained using accurate Quality Year (inception cohort)? or matching)? confounders? studied? cross-over)? on potential confounders? follow-up? methods? ratingBindman Unclear. No. Unclear. No for health status Unclear. Yes. Yes. Yes. Fair1990 Attempted to enroll all patients Differences between Patient surveys (self-reported). Age, sex, race, work status, 88% of Shasta Self-reported health status

meeting inclusion criteria. 72% groups on demographic at baseline (just Unclear for mortality insurance status, number of chronic residents and 84% of via the Medical Outcomes of those in Shasta and 65% and clinical risk factors. after hospital (from vital statistics conditions, and baseline response. comparison county Study Short Form (MOS-SF): of those in the comparison closed) and 1 year registry and family residents completed pain, mental health, physical county were contacted and later. members, but no baseline surveys; 87% function, health perception, enrolled; those who were information on blinding and 89% of those were social function, role function. not contacted were likely to of data collectors). followed up. Mortality information from have moved (and were thus family members and Vital ineligible). Statistics Registry.

Brownell Unclear. Unclear. Unclear. Unclear. NA. No. Unclear. Unclear. Poor1999 Gives inclusion criteria and Categorized Mortality and Age and sex only. No information on Does not report if outcomes

time frame, but not clear if all patients by readmission from missing data. mortality and readmission eligible were included. their place of hospital discharge were prespecified.

residence, not data; no information where they on blinding of data received their collectors.care.

Buchmueller Unclear. No. Unclear. Unclear. NA. Unclear. Unclear. Yes. Fair2006 Survey response rates NR. Demographic and Calculated Death reports from For health outcomes (from surveys): No information on LA County Health Surveys for

socioeconomic changes in travel California’s DHS, but no income, health insurance coverage, missing data. health status and zip code differences between distance using zip information on blinding health status, neighborhood (geographic area); death groups. code data. of data collectors. characteristics such as number of reports from California DHS.

Accuracy of community health clinics in a zip confounder code and city-level unemployment ascertainment rates.unclear due to For deaths: controls for total use of self-report deaths, deaths by homicide and data without age distribution of deaths, number validation. of health clinics in a zip code

year; no adjustment for SES and demographic differences.

Curtis Unclear. Unclear. Yes. Unclear. NA. No. Unclear. Unclear. Fair2005 All patients meeting criteria Similar on most Trained research No information on Different patient No control for confounders. No information on Used prespecified quality

were studied, but not all sites characteristics except nurses collected blinding of data groups compared. missing data. of care indicators specific contributed to assessment of for differences in information from collectors. to diagnosis, but indicators quality for each area of care. smoking history medical records. were chosen partly based on

(outside St. John’s availability of collectible data.region) and prior congestive heart failure (St. John’s region).


119CONTENTS 34

Did the article Did the study maintain

Did the study attempt to Were the groups use accurate Were outcome comparable Was there enroll all (or a random comparable at methods for assessors and/ groups (report acceptable sample of) patients baseline on key ascertaining or data analysts attrition, differential loss to Were outcomes pre-meeting inclusion criteria, prognostic factors exposures blinded to the contamination, Did the study perform follow-up and or specified and defined, and

Author or a random sample (e.g., by restriction and potential exposure being adherence, and appropriate statistical analyses overall high loss to ascertained using accurate Quality Year (inception cohort)? or matching)? confounders? studied? cross-over)? on potential confounders? follow-up? methods? ratingHemmelgarn Yes. No. Yes. Unclear. NA. No. Unclear. Yes. Poor2001 Stated inclusion criteria, Differences between Used ICD-9 codes No information on Different patient Controlled for age, sex, urgency of No information on Calgary Regional Health

screened all discharge records groups on several from hospital blinding of data groups compared. admission, prior procedures, and missing data. Authority administrative within stated time period. clinical risk variables. discharge data. collectors. comorbidities. But no control group; hospital discharge data.

did not assess the potential impact of concurrent events; unable to determine whether any observed changes exceeded what would be naturally expected over time.

Hsia Yes. No. Yes. Unclear. NA. Yes. Unclear. Yes. Fair2012 Differences between No information on Different patient Age, race or ethnicity, sex, and No information on Patient-level data from

groups on several blinding of data groups compared. insurance status, comorbidity index missing data. the California Office of clinical risk variables, collectors. to adjust for baseline mortality risk, Statewide Health and in race, and insurance case mix of each hospital to control Planning Development Patient status. for hospital-level differences in Discharge Data.

patient acuity. Use data from state to document which hospitals had ED closures.Data on hospital-level characteristics from hospital financial and use reports.

Liu Yes for mortality. No. Yes. Unclear for mortality NA. No. Unclear. Yes for mortality Poor2001 Used database containing Differences between (blinding not reported). Different patient Controlled for age and sex only; No information on (Saskatchewan Ministry of

all overnight hospitalizations groups on several No for health status groups compared. matched on some variables; no missing data. Health administrative data). during the time period. clinical risk variables. (self-reported). control for comorbidities. Unclear for health status Unclear for telephone survey. (patient surveys and Random sample, but more interviews).respondents were female.

Rosenbach Unclear. Unclear. NA. Unclear. NA. No. Unclear. Unclear. Poor1995 Gives inclusion criteria and All had Medicare, but No information on Different patient Controlled for age and sex only. No information on Mortality, but data source not

time period, but not clear other characteristics blinding of data groups compared. missing data. reported.if all eligible hospitals were not reported. collectors.included.

Shen Unclear. Yes. Unclear. Unclear. NA. Yes. Unclear. Yes. Fair2012 Stated inclusion and exclusion MedPAR No information on Controlled for demographics and No information on Mortality from MedPAR

criteria, screened all discharge records but no blinding of data clinical risk factors. missing data. records. records within stated time information on collectors.period meeting criteria but did data collectors. not report number eligible and enrolled.

Evidence Brief: Effects of Small Hospital Closure on ... · 2003;9(2):59-71. 5. Bazzoli GJ, Lee W,...

Documents

Transcript of Evidence Brief: Effects of Small Hospital Closure on ... · 2003;9(2):59-71. 5. Bazzoli GJ, Lee W,...